This paper provides a brief description of Epilepsy and its impact upon cognitive functioning. Particular reference is made to memory in individuals affected, and to the prognosis for memory


It is noted (by Aicardi 1992) that a diagnosis of epilepsy may evoke a range of myths and prejudices that have been associated with seizure disorders. The implication is to provide maximal information to all those concerned with the child in respect of the likely impact of the condition and the needs that may be linked to it.

For example, it should be pointed out that epilepsy is not a disease in its own right, and that seizures are just one form of symptoms of various types of brain disfunction of which some are quite benign.

Further, it must be recognised that epilepsies are by no means necessarily life long conditions and that it is inappropriate to fear that brain tumours or other serious brain disorders are a common cause of the epileptic symptoms.

Epilepsy is a disorder of the chemical messages which pass between brain cells, and occurs when groups of cells begin to act together in a concerted and repetitive way so that volleys of messages spread out from these cells to the rest of the brain. These bursts of activity, which give off a characteristic electrical pattern, disrupt normal brain activity.

The epileptic activity may remain localised in the original site, or may spread to the whole of the brain (or large areas of it).

The effects will depend on a number of factors ... where the activity starts, where it spreads to, whether or not it interferes with the mechanics of consciousness, its duration, and which brain mechanisms it interrupts.

Temporal lobe epilepsy [where the sequences originate in the temporal lobes] may be the single commonest form of epilepsy.

Scarring of the brain is the commonest cause of temporal lobe epilepsy. This scarring may be the result of head injury or previous infections in the brain, but the most common cause is previous deprivation of the oxygen supply of that part of the brain, often in infancy and childhood. This, in turn, may stem from birth injury or a severe infection, but most commonly is an outcome of convulsions linked to high body temperature.


Behavioural and Educational Issues

The majority of children with epilepsy do not experience intellectual disabilities, nor do they show any behavioural symptoms or abnormalities. However, between 20 and 30 per cent of children do show intellectual or behavioural problems at least partially as a result of damage to the central nervous system that also is responsible for the epileptic seizures.

Various educational and behavioural difficulties are observed in children with epilepsy but without learning difficulties. For example, the Isle of Wight study (Rutter et al 1974) indicated that the risk of psychiatric disorder among children with epilepsy is four times greater than among the general child population (and the risk is greater still when the epilepsy accompanies other neurological disorders).

The reason for the emotional and behavioural problems are wide ranging and may include abnormal brain activity, the impact of medication, and environmental/social factors.

In respect of the brain dysfunction, it is the precise location of the damage underlying the epilepsy which will determine the nature and degree of the problem such as Attention Deficit Disorder..., while anti-epileptic drugs can markedly influence behaviour in terms of level of arousal, attention, and memory.

Nevertheless, a major contribution to emotional (and, thus, to behavioural) problems comes from social factors, notably the stigmatisation that is linked to ignorance or prejudice.

Where the epilepsy takes the form of temporal lobe seizures, it is common to find Hyperactivity (A.D.D. & H). For example, Lindsay et a (1979) found an incidence of 30% within a large British sample; and it is noted that symptoms of this kind tend to be associated with relatively low IQ scores (median 70) and with social limitations in adult life.

Anticonvulsant medication constitutes a further risk in terms of emotional and behavioural symptoms. For example, phenobarbitone may cause a range of symptoms among children ... somnolence, irritability, restlessness, and learning impairment.

Given the possible side effects of medication, its usage should be limited among children. For example, it is recommended that children who have had two or more seizures may be candidates for drug treatment, but, even here, treatment may be withheld for those children with infrequent seizures that are of limited expression or which are not disruptive to routines (i.e. which occur at "manageable" times).

The choice of drug(s) used is necessarily arbitrary since neither the positive effects nor the undesirable effects are wholly predictable. For example, one might avoid phenobarbitone in the first instance because of the likelihood of behavioural disturbance. [It is also the case that the side effects have all the greater impact by making physically obvious the differences between children with epilepsy and controls, and confirming the prejudices that these children are indeed "special".]

Similarly, treatment would be initiated with a single drug thus to minimise the number and interaction of side effects. Where there is limited positive benefits, substitution by other drugs should precede the use of combinations of drugs, but one would avoid linking two drugs whose side effects are similar.

Treatment of the behavioural difficulty involves a need to analyse the various factors that contribute to the problem ... the epilepsy itself, the brain dysfunction, the drug effects, constitutional factors, and psycho social factors ... and to determine whether the epilepsy or the behavioural difficulty has the major impact upon the child (and his family and significant others).

For example, the emphasis in the past may have been upon drug and dietary treatment and the only outcome criterion was the (change in) frequency of seizures, with little or no attention given to emotional or behavioural outcomes. Parents and teachers may feel that the latter are more difficult to face than the epilepsy itself and it can happen that alleviating some of the affect or behavioural products may reduce the frequency or severity of the fits.

It is also emphasised that where children suffer treatment-resistant epilepsy and may receive surgical treatment, it is appropriate to arrange a neuropsychological assessment before as well as after the surgery. The outcome of the surgical intervention cannot be comprehensively evaluated unless there is a clear recognition of the pre-surgery problems ... cognitive, behavioural and psycho social, i.e. the possible "costs" of surgery (in terms of a time lag in respect of re-establishing certain functions, such as long term memory storage and retrieval) may be outweighed by the benefits in terms of reducing the range of symptoms associated with the epilepsy and of minimising the impact of negative social factors in terms of peer or carer anxiety or even rejection.

It is reported that some epilepsies are totally unresponsive to drug treatment such that surgery has to be considered at an earlier stage than would have been customary in the recent past.

Most surgery techniques are intended to remove the tissue that is responsible for the initiation and spread of the seizure discharge, and would be considered when the seizures are severe, when evidence points to a localised brain area, and when there is considered to be an ability to remove the tissue without producing unacceptable effects upon performance. Temporal lobe resection is the most commonly performed operation.

Under achievement at school is significantly more common in children with epilepsy than in the general population (Seidenberg et a 1986). This reflects, in part, the impaired intellectual abilities in some children ... but it can also be the case that children with epilepsy may acquire intellectual difficulties for which the epileptic condition itself, the treatment, and social factors may all be partly responsible There is consistent evidence (e.g. Loiseau 1990) that a disproportionate number of young people with epilepsy do not continue far, or at all, in further education; or they seek jobs below their capability level ... and it is suggested that this reflects low expectations from parents and others or low self-expectations.

Further, Loiseau et a (1983) have shown that children with absence epilepsy do less well socio-professionally than would be expected from their measured ability and relative benignity of their condition. One theory concerns the possible interference of the epileptic discharges, albeit of a mild level, upon learning or other mental processes where attention and concentration are relevant. Whatever the truth of this may prove to be, there are two possible implications:

(i) What might be regarded as problems of memory (whether storage, consolidation, or retrieval) may actually be a matter of initial attention to the stimulus.

(ii) The considerable learning and behavioural difficulties of many epileptic children would hardly be explained by the effects of interference, and one may need to postulate psycho-social factors and specific neural deficits.

In this regard, it may be relevant to quote Vermeulen et a (1993) whose findings were that there is no simple relationship between subjective evaluations of memory and test performance. Patients' memory complaints do not accurately predict disturbances that can be measured on standard neuropsychological measures; and these complaints may be related more to neuroticism or psycho social difficulties.

Epilepsy and Memory

Inhibited memory functions may be the most commonly reported problem associated with epilepsy... for example, Loiseau et al (1988) commented that memory deficits merit special attention since the people concerned seek help for these more frequently than for other impairments.

An immediate implication, therefore, is that in cases of epilepsy not amenable to drug treatment the possible "costs" of surgery... in terms of possible impact upon memory ... need to be set against the possibly greater costs of the untreated condition with its continual impact upon memory (including the "social" problems in respect of remembering names or events).

Assessments of the effects of neural damage are complex, because the behaviour of the individual with such damage will be determined by the C.N.S. systems that remain intact. Thus, if deficit in learning or memory is observed, it cannot be assumed that the damaged part of the brain is irrelevant to that learning or memory function. Further, one may believe that certain learning capacities are undamaged simply because measuring techniques are relatively crude and do not register the deficit.

The neural mechanism responsible for learning may contain "elements in parallel" so that the destruction of one of them does not necessarily impair the whole system; and much of the central nervous system is composed of "series networks" which can be interrupted at many different levels.

The complexity of the question of memory storage and the "localisation" of memory may be demonstrated by the way in which memories may survive even large scale brain damage; and in which specific memories, perfect in every detail, may be evoked by electrical stimulation in a variety of cortical sites.

Nevertheless, the implication may again suggest that, in cases of severe epilepsy, the decision to provide surgical treatment may not be as drastic as non-surgical intervention given the evidence that the seizures have a marked impact upon the memory consolidation process and that memory is not located in specific areas of the brain but may be stored via increased neural connections or facilitated chemical transmissions across a relatively wide area of the brain.

Thompson (1991) reviews research evidence which gives objective support to the clinical impressions of memory disturbance among patients with epilepsy ... and it would appear that individuals with problems of seizures originating in the temporal lobes are at the greatest risk.

Among the epileptic samples with temporal lobe seizures, comparisons have been made between those with right and those with left lobe "sites" with the general outcomes (e.g. Delaney et al 1980) that individuals with problems focused on the left do less well on verbal memory tests, and those with right lobe foci do less well on non-verbal memory tests.

The memory disturbance among epileptics may arise from a number of sources.

Firstly, the illness or injury which gave rise to the epilepsy may have a direct impact upon memory and other functions, i.e. the memory problem may reflect brain damage rather than the epilepsy per se.

It is found that patients with known brain pathology exhibit greater memory problems than do those for whom the cause of the epilepsy is unknown.

Further, the seizures themselves may underlie memory disturbance... especially when their origin is in the temporal lobe. However, it must be noted that dysfunction in different parts of the temporal lobe may give rise to various types of memory disorder (and awareness of the variation may depend upon the nature of the tests available).

In general, the combination of early age of onset and long duration of seizures has been associated with the greater memory impairments (e.g. Britain 1980). However, it has also been argued by Loiseau et a (1983) that age of onset during the adolescent years may place an individual at greater risk for learning and memory difficulties. Frequency of seizures is also directly related to greater disruption of memory.

Another factor thought to influence cognitive functioning, including memory, is the occurrence of "sub clinical" epileptic discharges. Such discharges may interfere with short term memory or the consolidation process, but established memories are left undisturbed.

Although the issue of attention has already been mentioned, one would usefully underline this point by quoting Weiskrantz and Koella (1981) who state that it is impossible to study memory without a subject who can see and hear and attend and respond. In any single test, it may not be possible to know which of the capacities is impaired.

There is the additional complication that conditions that affect an individual's mood, ability to recognise reinforcement, alertness, and motivation can all influence that performance.

Attention may be critical in that where there is difficulty in maintaining attention there will be limited learning. Evidence exists for attentional problems in some subgroups of epileptics, and that individuals with generalised seizures are most at risk .. this may explain why some studies have not found significant differences between patients with temporal lobe epilepsy and those with generalised epilepsy on measures of memory.

Organisational skills may also be significant in that certain cognitive "styles" among epileptic individuals may inhibit acquisition of material and lead to poor scores on memory tests. Alderkamp (1983) focuses particularly upon those with impulsive reactions and those with rigid thinking styles.

Meanwhile, one should remain aware of the concept of "state dependent learning" whereby the material learnt while the person is in a particular mood, level of arousal, or drug affected state may be most effectively recalled in a similar situation.

In any event, the five most frequently reported types of day by day memory failures are described as follows:

1. Finding a word is on the "tip of the tongue"

2. Having to go back to check one has done things

3. Forgetting one was told something yesterday or a few days ago

4. Forgetting people's names

5. Losing things around the house.

Treatment, Surgery and Memory

The effects of surgery upon memory will depend upon the nature of existing deficits, the location of the surgery and the existence of other impairment.

For example, Thompson (opp. cit) reviews the evidence and describes how a severe memory disorder has been observed following a resection of one temporal lobe when the contra lateral lobe is not intact. Where the contra lateral lobe is not impaired, then temporal lobectomy will exacerbate existing memory difficulties.

Further, a resection of the dominant lobe generally impairs verbal learning and memory, and resection of the non-dominant lobe generally impairs non-verbal learning and memory.

It may be argued that greater impairments result from dominant lobe resection, but this may be a reflection of the relative significance of verbal memory in day to day experience, such that the impact of impairment in this respect is more noted.

The more selective the surgery technique, the fewer deficits in memory may be anticipated. On the other hand, surgery may have a negative impact if seizures are not better controlled as a result of the operation

More recent studies suggest that an improvement in memory functions subserved by the contra lateral temporal lobe can occur following the operation... as if the memory functions of the unaffected hemisphere had been somehow inhibited by the seizures in the resected hemisphere.

Consensus suggests that impaired memory measured after the operation remains fairly permanent although some findings may suggest some recovery 3-7 years after treatment, with young adults showing the earliest recoveries.

It is found that frontal lobe surgery may also impair memory, especially where a memory task requires monitoring of sequences in time (see Smith and Miner 1984), although this kind of finding might not be surprising given the evidence for the organisational functions of this part of the brain.

Continuing the theme of localisatlon of neural deficits, Dodrill et al (1993), suggest that there is a greater probability of seizure relief following surgery when the site of the surgery and its localisation of dysfunction as measured by neuropsychological tests are matched.

When the test results indicate that the dysfunctions are non-lateralisable, then the outcome of surgery in terms of seizure relief may be poor.

The authors also quote findings where cases with good outcome were more likely to have right hemispheric deficits than left-side deficits ... possibly because of a tendency to make larger resections on the right and thereby to gain greater relief from seizures.

Measures of how well an individual performs on tests of mental abilities may predict the extent of seizure relief after surgery.

The first evidence for this was produced by Wannamaker and Matthews (1996) in their finding that the persons less impaired before. surgery had the best prognosis although comparable studies have not replicated this outcome.

Similarly, some research has identified a link between greater pre-operative measured intelligence and the greater probability of seizure relief; while other studies have not identified a relationship.

While overall indicators of mental abilities seem to have restricted usefulness, scores on particular items may be more useful,

For example, two measures of ability (WAIS Digit Span, and Marching Test) and two adjustment measures (MMPI Hysteria Scale and Paranoia Scale) were found to be predictive of seizure relief beyond that afforded by EEG or other medical variables.

putting together the psychological data and EEG data gives as accurate prediction whether or not an individual would experience at least a 75% reduction in seizure frequency after surgery.

In respect of pre and post surgery intellectual testing (using the Wechsler Scales) the available findings ... such as those of Augustine and Novelly (1981)... indicate that individuals who have temporal lobe surgery may show some modest gains in intelligence scores, especially if the resection is on the side not associated with speech, and if they become seizure-free. A note of caution should be introduced, however, in respect of the matter of practice effect which itself would be expected to lead to some positive gains anyway.

Returning to the theme of memory and surgery, one notes the general finding that global memory deficits are very rare outcomes of temporal lobe surgery, but material-specific deficits continue to be observed (see, for example, Chelune 1991). The risk of declines in verbal learning and recall ability following left-side surgery has been consistently identified, while post-operative declines in non-verbal abilities following right-side surgery are not consistent. The problem, as noted by Dodrill et al (opp, cit.) is that, whatever the trend in group outcomes, it may be very difficult to identify the potential risk to memory among individuals.

There is, however, converging evidence that, for both adults and children, the greatest risk of memory loss occurs for those people who appear to have the most intact pre-operative memory function. It appears that the critical neural site is the hippocampus where this is functioning well, the threat to memory from surgery is greater; where the hipocampus is not functioning effectively, the surgery will result in less detectable memory loss.

Wolf et al (1993) recognise that decreased memory or learning efficiency may follow left temporal lobe surgery, but investigated whether the acquired deficit is related to the size of the resection. Their results showed similar outcomes of both limited and extended resections, whether mesial or lateral; and it was argued that the risk of cognitive impairment depended more on age at seizure onset than on the extent of resection (i.e. there appeared to be a negative correlation between the subject's age and the extent of cognitive change).


Any attempt to explore memory functions in epilepsy is complicated by the many factors which can influence efficiency, but also by the multi-faceted nature of memory itself... i.e. sensory memory, recognition memory, short-term memory, working memory, long-term memory, episodic or autobiographical memory, verbal/non-verbal memory, and "meta memory".

Disorders of memory may occur in one or more memory systems, and the defect may be within any one of a number of mechanisms .... encoding, storage, or retrieval.

Epilepsy refers to a heterogeneous group of disorders, and an equally wide variety of types of memory disorder may be encountered.

The difficulties may be the more complex when the memory disturbance is secondary to a general effect such as impaired concentration; and because epilepsy is not a static condition, the nature and severity of memory problems experienced by individuals may change with time.

M.J. Connor June 1994


Aicardi J 1992 Diseases of the Nervous System in Childhood London. Mackeith Press

Aldenkamp A 1983 Epilepsy and Learning Behaviour in Parsonage et a (eds) Avances in Epileptology; XlVth Symposium. New York. Raven Press

Augustine E and Novelly R 1981 Memory and IQ Correlates of Seizure Control Following Anterior Temporal Lobectomy Epilepsia 48. 317-326.

Britain H 1980 Epilepsy and Intellectual Functions in Ku hg et a (Ads) Epilepsy and Behaviour. Lisse. Swets

Chelune G 1991 Using Neuropsychological data to Forecast Post Surgical Cognitive Outcome. In Luders H (ed.) Epilepsy Surgery New York Raven Press

Delaney R, Rosen A, Mattson R, Novelly R 1980. Memory Function in Focal Epilepsy. Cartex 16.103-117.

Dodrill C, Hermann B, Rausch R, Chelune G and Oxbury S 1993 Neuropsychological Testing for Assessing Prognosis Following Surgery for Epilepsy in Engel J (ed.) Surgical treatment of the Epilepsies New York Ravel Press.

Lindsay J, Ounsted C, Richards P 1979. Long-term outcome in Children with Temporal Lobe Seizures Developmental Medicine and Child Neurology 21. 630-696

Loiseau P, Pestre M, Dartigues J, Commenges D, Barberger-Gateau C, Cohadon S 1983 Long term prognosis in two forms of childhood epilepsy Annals of Neurology 13 642-648

Loiseau P, Strube E, Broustet D, Battelochi S, Gomeni C, Morselli P, 1983. Leaming Impairment in Epileptic Patients Epilepsia 24. 183-192.

Loiseau P, Strube E, Signoret J 1988 Memory and Epilepsy in Trimble and Reynolds (eds) Epilepsy, Behaviour and Cognitive Function Chichester,. Wiley.

Olsson I 1990 Absence Epilepsy in Swedish Children M.D. Thesis Goteburg (ISBN 91-628-0148-1)

Rutter M, Graham P, Yule W 1970. A Neuropsychiatric Study in Childhood Clinics in Developmental Medicine 35136. London Spastics International Medical Publications with Heinemann Medical.

Seidenberg M, Beck N, Geisser M, Giordoni B, Sackellares J, Beren S, Dreifuss F, Boll T 1986. Academic Achievement of Children with Epilepsy Epilepsia 27. 753-759.

Thompson P 1991. Memory Functions in Patients with Epilepsy. Advances in Neurology Vol.55 (eds Smith et al). New York Raven Press

Vermeulen J, Aldenkamp A, Alpherts W 1993 Memory Complaints in Epilepsy Epilepsy Research 15(2) 157-170.

Wannamaker B and Matthews C 1976 Prognostic Implications of Neuropsychological Test Performance for Surgical Treatment of Epilepsy Journal of Nervous Mental Disorders 163 29-34.

Weiskrantz L and Koalla W 1981. Highlights in Presented Papers on Clinical Aspects of Memory Function. Acta Neurol Scand 64 78-82.

Wolf R, Ivack R, Hirschorn K, Sharbrough F, Cascino G, and Marsh W 1993. Neurocognitive Efficiency Following Left Temporal Lobectomy. Journal of Neurosurgery 79(1) 76-83.

This article is reproduced by kind permission of the author.

© Mike Connor 1994.

Back to NAS Surrey Branch Welcome Page